Chapter 5: Revisability and Conceptual Change

Perhaps the most famous attack on Carnap’s logical empiricism is W.V. Quine’s article “Two Dogmas of Empiricism”. In the article, Quine argues against the analytic/synthetic distinction that Carnap employs, as well as against his sometime verificationism. The article is widely regarded as much more than an attack on logical empiricism, however. It is often seen as the most important critique of the notion of the a priori, with the potential to undermine the whole program of conceptual analysis. In this chapter, I address Quine’s most influential arguments in “Two Dogmas”. I do this in part for defensive reasons and in part for constructive reasons. Defensively: Quine’s arguments might be thought to undermine my frequent appeals to the a priori, so addressing these arguments helps to support those appeals. Constructively: addressing Quine’s arguments in the spirit of the scrutability framework helps to bring out some of the power of that framework. For example, it helps us to analyze notions of meaning in a broadly Carnapian manner, and it helps us to understanding the relationship between rationality and conceptual change. I will address Quine’s article construed as a critique of the notions of analyticity and apriority. I am more inclined to defend the notion of apriority than the notion of analyticity, so I will focus more on the former, but the response that I will develop can be used to defend either notion from Quine’s arguments. I will focus especially on the most influential part of Quine’s article: the arguments in the final section concerning revisability and conceptual change. In addressing these arguments, I will adopt a line of response grounded in Carnap’s underappreciated article “Meaning and Synonymy in Natural Language”. Carnap’s article offers an approach to issues about meaning that is highly congenial to the scrutability framework. I will argue that an analysis inspired by this article, when conjoined with tools drawn from the scrutability framework and from Bayesian confirmation theory, provides just what is need to reject Quine’s argument.